API

Cache any python object to file using improved pickling.

Classes:

Name	Description
AnyCache	Cache for python objects.

Functions:

Name	Description
anycache	Decorator to cache result of function depending on arguments.
get_defaultcache	Return unlimited default AnyCache instance.

AnyCache

Cache for python objects.

Other Parameters:

Name	Type	Description
cachedir	Optional[Path]	Directory for cached python objects. AnyCache instances on the same cachedir share the same cache.
maxage	Optional[timedelta]	Maximum cache item age in seconds.
maxsize	Optional[int]	Maximum cache size in bytes. None does not limit the cache size. 0 disables caching. It the maximum size is smaller than the last cached object, this object is kept. During object write the cache size might be larger than maxsize. At maximum twice as large as the maximum object size.

The AnyCache instance mainly serves the AnyCache.anycache method for caching the result of functions.

>>> from anycache import AnyCache
>>> ac = AnyCache()
>>> @ac.anycache()
... def myfunc(posarg, kwarg=3):
...     print("Calcing %r + %r = %r" % (posarg, kwarg, posarg + kwarg))
...     return posarg + kwarg
>>> myfunc(4, 5)
Calcing 4 + 5 = 9
9
>>> myfunc(4, 5)
9
>>> myfunc(4, 2)
Calcing 4 + 2 = 6
6

The cache size is returned by AnyCache.size.

>>> ac.size
10

The cache size can be limited via maxsize. A maxsize of 0 disables caching.

>>> ac.maxsize = 0
>>> myfunc(4, 5)
Calcing 4 + 5 = 9
9

The cache is preserved in this case, and needs to be cleared explicitly:

>>> ac.size
10
>>> ac.clear()
>>> ac.size
0

Methods:

Name	Description
anycache	Decorator to cache result of function depending on arguments.
is_outdated	Return True if cache is outdated for func used with args and kwargs.
remove	Remove cache data for func used with args and kwargs.
get_ident	Return identification string for func used with args and kwargs.
clear	Clear the cache by removing all cache files.

Attributes:

Name	Type	Description
cachedir		Cache directory use for all cache files.
size		Return total size of all cache files.

cachedir property writable

cachedir

Cache directory use for all cache files.

AnyCache instances on the same cachedir share the same cache.

size property

size

Return total size of all cache files.

anycache

anycache(depfilefunc=None)

Decorator to cache result of function depending on arguments.

Other Parameters:

Name	Type	Description
depfilefunc	Optional[Callable]	Dependency file function (see example below)

Example:

>>> from anycache import AnyCache
>>> ac = AnyCache()
>>> @ac.anycache()
... def myfunc(posarg, kwarg=3):
...     print("Calcing %r + %r = %r" % (posarg, kwarg, posarg + kwarg))
...     return posarg + kwarg
>>> myfunc(2, 5)
Calcing 2 + 5 = 7
7
>>> myfunc(2, 5)
7

File I/O is not tracked by the decorator. Instead a function needs to be implemented, which returns the paths of the files, which influence the function result. The depfilefunc is called with the function result and all arguments. The following example, depends on the path of the source code itself:

>>> def mydepfilefunc(result, posarg, kwarg=3):
...     print("Deps of %r + %r = %r" % (posarg, kwarg, result))
...     return [__file__]
>>> @ac.anycache(depfilefunc=mydepfilefunc)
... def myfunc(posarg, kwarg=3):
...     print("Calcing %r + %r = %r" % (posarg, kwarg, posarg + kwarg))
...     return posarg + kwarg
>>> myfunc(2, 7)
Calcing 2 + 7 = 9
Deps of 2 + 7 = 9
9

is_outdated

is_outdated(func, *args, **kwargs)

Return True if cache is outdated for func used with args and kwargs.

Example:

>>> from anycache import AnyCache
>>> ac = AnyCache()
>>> @ac.anycache()
... def myfunc(posarg, kwarg=3):
...     print("Calcing %r + %r = %r" % (posarg, kwarg, posarg + kwarg))
...     return posarg + kwarg
>>> ac.is_outdated(myfunc, 2, 5)
True
>>> myfunc(2, 5)
Calcing 2 + 5 = 7
7
>>> ac.is_outdated(myfunc, 2, 5)
False

remove

remove(func, *args, **kwargs)

Remove cache data for func used with args and kwargs.

>>> from anycache import AnyCache
>>> ac = AnyCache()
>>> @ac.anycache()
... def myfunc(posarg, kwarg=3):
...     print("Calcing %r + %r = %r" % (posarg, kwarg, posarg + kwarg))
...     return posarg + kwarg
>>> myfunc(2, 5)
Calcing 2 + 5 = 7
7
>>> ac.remove(myfunc, 2, 5)
>>> myfunc(2, 5)
Calcing 2 + 5 = 7
7

Removing non-existing cache entries is not an error:

>>> ac.remove(myfunc, 2, 5)
>>> ac.remove(myfunc, 2, 5)

get_ident

get_ident(func, *args, **kwargs)

Return identification string for func used with args and kwargs.

Example:

>>> from anycache import AnyCache
>>> ac = AnyCache()
>>> @ac.anycache()
... def myfunc(posarg, kwarg=3):
...     print("Calcing %r + %r = %r" % (posarg, kwarg, posarg + kwarg))
...     return posarg + kwarg
>>> @ac.anycache()
... def otherfunc(posarg, kwarg=3):
...     print("Calcing %r + %r = %r" % (posarg, kwarg, posarg + kwarg))
...     return posarg + kwarg
>>> ac.get_ident(myfunc, 2, 5)
'19044d3869955fa79d7f3db8fcdc5af84b3f55c0bdab2b4aee1bb21e1a9856c9'
>>> ac.get_ident(myfunc, 2, 6)
'1885e09f9898a1f1bd186f052d0e810693faa14b70e7b6b22de61b90c8171427'
>>> ac.get_ident(otherfunc, 2, 5)
'9b8aea26422999aaa7aed0fdb4d5145fd33b87de20f322cf175997e9b1835158'

clear

clear()

Clear the cache by removing all cache files.

anycache

anycache(cachedir=None, maxsize=None, depfilefunc=None)

Decorator to cache result of function depending on arguments.

This decorator uses one unlimited global cache within one python run. Different anycached functions have different cache name spaces and do not influence each other.

To preserve the cache result between multiple python runs, use an AnyCache instance with a persistent cachedir.

Other Parameters:

Name	Type	Description
cachedir	Optional[Path]	Directory for cached python objects. AnyCache instances on the same cachedir share the same cache.
maxsize	Optional[int]	Maximum cache size in bytes. None does not limit the cache size. 0 disables caching. It the maximum size is smaller than the last cached object, this object is kept. During object write the cache size might be larger than maxsize. At maximum twice as large as the maximum object size.
depfilefunc	Optional[Callable]	Dependency file function (see example below)

Example:

>>> from anycache import anycache
>>> @anycache()
... def myfunc(posarg, kwarg=3):
...     print("Calcing %r + %r = %r" % (posarg, kwarg, posarg + kwarg))
...     return posarg + kwarg
>>> myfunc(2, 5)
Calcing 2 + 5 = 7
7
>>> myfunc(2, 5)
7

File I/O is not tracked by the decorator. Instead a function needs to be implemented, which returns the paths of the files, which influence the function result. The depfilefunc is called with the function result and all arguments. The following example, depends on the path of the source code itself:

>>> def mydepfilefunc(result, posarg, kwarg=3):
...     print("Deps of %r + %r = %r" % (posarg, kwarg, result))
...     return [__file__]
>>> @anycache(depfilefunc=mydepfilefunc)
... def myfunc(posarg, kwarg=3):
...     print("Calcing %r + %r = %r" % (posarg, kwarg, posarg + kwarg))
...     return posarg + kwarg
>>> myfunc(2, 7)
Calcing 2 + 7 = 9
Deps of 2 + 7 = 9
9

get_defaultcache

get_defaultcache()

Return unlimited default AnyCache instance.